The present investigators have noted that radiologists and otolaryngologists who have entered the field in recent years have a significantly higher rate of multiple myeloma than U.S. white males in general. It has been proposed that exposure to repeated low-doses of radiation would have a simulatory effect on the immune system and thus result in plasma cell dyscrasias. Confounding variables such as exposure to infections and chemicals could influence this risk in physicians. Tumors of the brain occur with unusual frequency among internists with rates significantly higher than those seen in dentists. Very little is kn wn about the etiology of these tumors but it has been noted in the total U.S. County Study that the mortality for brain tumors and multiple myeloma are correlated. The proposed research project would study the etiology of these two cancers and their correlation in physicians, a relatively homogeneous population distributed across the U.S. This group is unique in having a high probability of exposure to repeated low-dose radiation and infections and go d medical histories. The design is a case-control study with an estimated 350 cases of each of the cancers and a stratified random sample of 700 deceased controls comparable by age, race, sex and year of death selected from the total physician deaths over 15 years. Equal numbers of living controls will also be selected. The risk factors identified in previous studies will be examined by a telephone interview to the next-of-kin, the physician's associates, or the physician if a living control. There will be special emphasis on determining the exact number of years of exposure to radiation and the estimated dose levels by year or dosimetry data. Questions concerning patient contacts, personal characteristics and medical history will be included. Information on residence, type of training and specialty are available from AMA records. A Multi-variate analysis of data will be used to determine association of each of these cancers with radiation, infection, or other factors.